The repository contains PyTorch implementation of the SimCLR self-supervised learning method.
It is adapted to be used on earth observation data from the BigEarthNet dataset/benchmark, with multilabel classification as downstream task (whether with a linear probe or finetuning). The repository also offers an implementation of a classifier for the same multilabel classification task as a standalone task, so that a direct comparison with the outcome of SimCLR's downstream task is easily and directly obtainable.
There is one script to run the job locally (anton_launcher)
and one adapted for Slurm-orchestrated HPC clusters (slurm_launcher).
The file spawner.py at the root (called by slurm_launcher)
also offer the option to spawn arrays of jobs locally within a Tmux session
(where each job spawned has its own window within the session).
When run locally, the scripts expect the dataset to be present locally uncompressed. When run with Slurm, it expects the dataset to be present on a accessible note compressed. These behaviors are modifiable in the scripts provided. The choices I made here were for my own convenience.
GDAL: might need to install gdal and/or libgdal on your system.
Has proven finicky to install.
Python version: >=3.10
Set up your Python environment as follows (ordering is important):
pip install --upgrade pip
conda install -c conda-forge gdal
pip install rasterio opencv-python tqdm numpy scikit-learn wandb tmuxp tabulate pyright ruff-lsp
pip install torch torchvision torchaudio